The simulation of marine operations, in particular of lifting operations, requires the modeling of the whole system (ship, cable and payload) along with a multibody theory, an appropriate hydrodynamic theory and cable's modeling. This paper discusses a new approach to achieve this type of simulation.

The multibody theory uses a robotics formalism and a direct dynamic algorithm based on recursive techniques for kinematic trees to solve the Newton-Euler equations. Each body is separated by a joint, either revolute or prismatic, to grant for a degree of freedom. The multibody system is composed of the ship, the cable (made of several elements) and the payload. The cable modeling is based on the same multibody approach. Each cable element is composed of three joints (two revolutes and one prismatic). Only the axial tension and the axial damping force are taken into account in the internal loads. There is neither bending nor torsion effects. We verified that this model matches exactly with the classical lumped mass theory in the case of a payload hanging on a crane. The multibody algorithm has been adapted to speed up the computation of the mechanical solver when there are cable elements. In order to lower or lift the payload, a method is used to dynamically add or delete elements of the cable according to a criterion.

Hydrodynamic loads are computed using a weakly non-linear potential flow solver, based on a weak-scattered hypothesis. The velocity potential is splitted into two parts, the incident and the scattered (perturbation) components. The weak-scatterer hypothesis assumes the perturbation part is small compared to the incident one and the free surface boundary conditions are linearized with respect to the incident wave elevation. This time-domain method allows simulating a lowering operation from the air to the seabed and through the free surface. Unsteady effects due to the change of hydrodynamic coefficients or the interactions between the ship and the payload can be taken into account. This is not possible with a linear potential flow solver. This tool was originally developed for a single floater. It has been extended to the cases with several floaters.

Moreover, these two solvers (mechanic and hydrodynamic) need to be coupled in order to perform the simulation. Indeed, the acceleration of each floater depends on the acceleration potential on the floater which depends itself on the acceleration of this body. The strategy to manage this tight coupling is presented.

Finally, a test case with a floater, a cable and a payload was performed and the results are presented in this paper.